Neural network scoring of rat sleep stages

Analysis and classification of sleep stages is a fundamental part of basic sleep research. Rat sleep stages are scored based on electrophysiological signals normally recorded from electrodes implanted epidurally and in the temporalis muscle. The process of scoring has generally been a labor and time intensive procedure involving the manual analysis of the time domain biological signals recorded on a stripchart. The authors' laboratory has developed a computer-based system capable of recording over a period of several days. This system allows for on-line manual scoring as well as providing summary statistics of the data. In order to reduce the labor and time involved in scoring lengthy recordings, a neural network approach to classifying sleep stages has been developed.